Provides a general and efficient tool for fitting a response surface to datasets via Gaussian processes. The dataset can have multiple responses and the fitted GP model can predict the gradient as well. The package is based on the work of Bostanabad, R., Kearney, T., Tao, S. Y., Apley, D. W. & Chen, W. (2018) Leveraging the nugget parameter for efficient Gaussian process modeling. International Journal for Numerical Methods in Engineering, 114, 501-516.
| Version: | 1.0.1 |
| Depends: | R (≥ 3.2.5), stats (≥ 3.2.5) |
| Imports: | lhs (≥ 0.14), randtoolbox (≥ 1.17), lattice (≥ 0.20-34), grDevices, graphics |
| Published: | 2018-06-03 |
| Author: | Ramin Bostanabad |
| Maintainer: | Ramin Bostanabad <bostanabad at u.northwestern.edu> |
| License: | GPL-2 |
| NeedsCompilation: | no |
| CRAN checks: | GPM results |
| Reference manual: | GPM.pdf |
| Package source: | GPM_1.0.1.tar.gz |
| Windows binaries: | r-devel: GPM_1.0.1.zip, r-release: GPM_1.0.1.zip, r-oldrel: GPM_1.0.1.zip |
| OS X binaries: | r-release: GPM_1.0.1.tgz, r-oldrel: GPM_1.0.1.tgz |
| Old sources: | GPM archive |
Please use the canonical form https://CRAN.R-project.org/package=GPM to link to this page.